This paper was attempted to examine the effect of temperature and rainfall on the productivity of two important crops-pigeonpea and cotton in Andhra Pradesh following panel data regression approach. Using the district level time series data, the yield of each of these two crops was regressed on maximum temperature during kharif, rainfall quantity and number of rainy days for the period 1990–2002, in a one-way and two-way fixed effect models of panel regression using Least Squares Dummy Variable Method. Comparison of district specific effects revealed statistically significant differences between districts in case of pigeonpea. The response coefficients for rainfall and number of rainy days were found significant at 5 per cent level. A positive impact on yield of pigeonpea at the rate of 1.9 kg/ha was observed for a 10 mm rise in rainfall. The number of rainy days was found to have a significant negative relationship with yield. In case of cotton, half of the major cotton growing districts differed significantly in mean yield levels. A significant reduction in yield at a rate of 13 kg/ha for every 1°C rise in the maximum temperature was observed. The yield of cotton was found to increase by 0.7 kg/ha for every 10 mm increase in the rainfall.